Document MMgKmo4exVanwZDjEEp6O1mNz

,, * 933 Buckaan Laboratories, Inc. P.o. SEP 30 1957 Swelling of Paint Films in Water VI. Effects of Different Oil or Oleoresinous Vehicles1 F. L. BROWNE Chemist, Forest Products Laboratory,* Forest Service, U. S. Department of Agriculture Discusses the effect of four oil or vornish vehicles on the behavior of free films of paints when soaked In distilled water for three days and then redried. The films were tested when 10 days old and after 15 days of artificial weathering. Measured were changes of density, absorption of water, swelling during absorption, shrinkage on redry ing, loss in weight while soaking, and loss in weight while weathering. h e f if t h p a p e r in this series (3 ume, area, and thickness when water T to 7)s described the effect of was absorbed, (6) of the contraction 44 different pigments on the behavior in volume, area, and thickness when of free films of single-pigment paints the films were redried, (7) of rite loss made with linseed oil when the films of soluble matter from the films while were soaked in distilled water for 3 they were soaking, and (8) of the days and then redried. The films were loss in weight while the films were tested when 10 days old and also after artificially weathered. an additional 13 days of artificial In the meantime, a number of pub weathering. lications on the swelling of films by This sixth paper describes similar other investigators have appeared (1, 9 studies of the effects of different oil to 11, 16 to 22). and oleoresinous vehicles. Unpigmented films of bodied linseed oil, an Vehicles and Paints Tested alkyd-resin vehicle recommended for Three paint vehicles were chosen: a making house paints, and a phenolic- heat-bodied linseed oil of viscosity Z-6; resin varnish were tested both before a commercial long-oil, pure alkyd-resin and after artificial weathering. The vehicle sold for use in "stain- and results were compared with previous blister-resistant" house paints; and a data for films of unbodied linseed oil. pure phenolic-resin varnish of 33- Single-pigment and multiple-pigment gailon length in linseed and tung oils. paints were made with each of the A fourth vehicle, refined but unbodied vehicles, and films with these pigments linseed oil, had already been tested were tested before and after weather extensively, as reported in previous ing. The tests included measurements: papers (4, 7). Driers in the unbodied (1) of changes in density during for and bodied linseed oils were lead and mation of films, (2) on free films after manganese naphthenates; in the alkyd- they were soaked in distilled water for resin vehicle, they were cobalt and cal three days, (3) on the redried films, cium naphthenates; and in the phe (4) on the absorption of water during nolic-resin varnish, they were cobalt soaking, (5) of the swelling in vol- and lead naphthenates. 1A contributed paper. 1 Maintained at Madison, Wis. in cooperation with the University of Wisconsin. * Numbers in parentheses refer to literature cited. Unpigmented films of the alkydresin vehicle and the phenolic-resin varnish were tested, and data for un pigmented films of unbodied and The Author; Frederick L. Browne received B.Chem. degree from Cornell U., Ph D. in colloid chemistry from V. of Wisconsin. He first joined the rorest Products Lab staff in 1918, helped develop casein glues for wood airplanes of world War 1. In l$2V-22 he was a National Research Fellow at Wisconsin. Since 1929. Dr. Browne bas edited the General and Physical Chemistry section of Chemical Abstracts. bodied linseed oil were available from a previous paper (4). All experimental paints were made at 30 per cent pigment volume (symbol p/nv 0.30). Each kind of pigment was of the same make and grade in all ex perimental paints reported in this ^Wth each of the vehicles, there were single-pigment paints made with basic carbonate white lead (symbol L), antimony oxide (symbol A), rutile titanium dioxide (symbol T), zinc oxide (symbol Z), ana magnesium sili cate (symbol X). There were also three mixed-pigment paints as follows: (a) a lead-zinc paint LZ,, (the sub script represents the percentage of the pigment symbolized in the total pig ment by volume), (a) an antimonyzinc paint, AZ,p and (c) a titaniumzinc paint, TZ,6. In the alkyd-resin vehicle, the titanium-zinc paint also contained magnesium silicate, TZltX,,. A fourth mixed-pigment paint, TA,,X,,, was also used with the alkydresin vehicle. The five pigments for the single pigment paints were chosen on the basis of the absorption and swelling properties previously found for un bodied linseed oil paints (7). Two were chosen for unusually low, one for very high, and two for moderately high absorption and swelling. The mixedpigment paints represented two com binations of low- and high-swelling pigments and two combinations of moderately high- and high-swelling pigments. Preparation of Free Films and Tests Clear vehicles and paints to be tested before weathering were spread on gummed paper by doctor blade and suction plate, and allowed to dry for 10 days. The free films were then stripped from the paper by immersion in cold water for about 13 minutes until the paper backing became wet and the dextrine gum softened. Vehi cles and paints to be weathered before testing were spread on tinplate, and Reprinted from the April, 1956, Forest Products Journal (Vol. VJ, No. 4), pages 152-159 Forest Products Research Society, P. O. Box 2010, University Station, Madison 5, Wisconsin allowed to dry for 10 days. They were then sprayed with distilled water three times an hour in a Weatherometer for 15 days, and simultaneously exposed to ultraviolet light from a flaming car bon arc enclosed in Corex D glass (1,200 kilowatt-hours of input of electric current). The tinplate speci mens were weighed before coating, after the coatings had dried and cured, and after the period of artificial weath ering, to determine the loss in weight of the coatings while weathering. Free films were then stripped for testing by amalgamation of the tinplate. Procedures for measuring absorp tion, swelling, and solubility were essentially those described previously (4). Test films were cut 1^ by 2)^ inches in size, scribed with reference marks for measuring changes in length and width with a traveling microscope, and dried in a desiccator over calcium chloride. The films were then measured for length and width, and weighed in air ancf under water to determine the initial weight and volume. Each film was submerged in a crystallizing dish under 250 cubic centimeters of dis tilled water and a weighted pad of glass cloth to prevent undue curling of high-swelling films. The films were soaked for 3 days at 70 F., and weighed under water. They were then sandwiched between microscope cover glasses for measurement of length and width, removed from the cover glasses and quickly blotted with paper towels to remove free water, ana weighed in air. Finally, the films were redried for 3 days in a desiccator, and then meas ured for length and width and weighed in air and under water. At each of the three test stages, be fore and after soaking and after redry ing, the data yielded the weight in air, W, the volume, V, and the length and width between reference marks. The reasonable assumption that the whole film swelled or shrank in the same proportion as the part between refer ence marks made It possible to com pute the swollen and redried lengths, L, and widths, B, of the entire film from the known initial dimensions. The following relations then apply at each stage of test: Area Q= L X B Thickness T = V/Q Density D = W/V Indicating the initial, swollen, and re dried stages by the subscripts o, s, and r, respectively, changes in the film usually can be calculated from the relations: Absorption of water, A, in per cent by volume = 100 (W, -- Wr)/v,, Increase in volume, AV,, in per cent = 100 (V. -- V,)/V,, Increase in area, AQ,, in per cent = 100 <Q. - Qr)/Q,, Increase m thickness, AT,, in per cent = 100 (V./Q. - V./Q0) Swelling efficiency, in per cent = 100 AV,V,,/A Decrease in volume, AVr, in per cent = 100 (V0 - Vr)/V. Decrease in area, AQr, in per cent = 100 (Q, - Qt)/Q Decrease in thickness, ATr> in per cent = 100 (Vr/Qr -- V#/Q,,) Solubility, S,, the loss in weight while soaking, in grams per 100 cubic centimeters = 100 (W,, -- Wr)Ve Solubility, Sj, the loss in weight while weathering, is similarly cal culated in per cent as 100 times the weight loss divided by V0. In the literature on swelling of paint films, the absorption of water has often been reported m percentage by weight for convenience, but the great variation in density of paints impairs even the relative significance of such measure ments. Swelling is by nature a volu metric process. There is, therefore, a growing practice of reporting absorp tion voliimetrically (2, 8, 10). Expres sion of solubility in percentage by weight is similarly questionable. As long as the density of the soluble in gredients remains unknown, the most useful method of reporting solubility for the present purpose seems to be the weight of dissolved matter per 100 cubic centimeters of film. Occasionally the volume and often the area of a redried film remain greater than they were before soaking, despite loss of soluble substances to the soaking water. If the increase is beyond the limits of experimental error, either there has been a disproportionally large shrinking in thickness or a decrease m film density on redry ing after soaking. In such cases, the volumetric swelling was taken as AV, = V, -- V,, and the areal swell ing as AQ, = Q, -- Q- Unweathered Films The effects of soaking on unweath ered films are reported in Table 1. Density: All four clear vehicles be came more dense after they had dried for 10 days. Although the density of the liquid vehicle varied from 0.93 for unbodied linseed oil to 1.02 for the phenolic-resin varnish (nonvolatile only), the 10-day-old films all readied the same density, 1.11. The increase in density was 19.4, 14.4, 11.0, and 8.8 per cent of the density of die liquid unbodied linseed oil, bodied linseed oil, alkyd-resin vehicle, and phenolicresin varnish, respectively. The density of all paints made with the four vehicles similarly increased during the 10-day drying period. For paints made with unbodied linseed oil, the percentage increases ranged from 4.7 to 11.3 per cent; for paints made with bodied linseed oil, from 1.3 to 6.5 per cent; for paints made with alkyd-resin vehicle, from 2.9 to 8.6 per cent; and for paints made with phe nolic-resin varnish, from 0.8 to 3.2 per cent. It was shown previously (7) for unbodied linseed oil paints that the increase in density is always less than that calculated on the assumption that the oil in a paint undergoes die same changes in weight, volume, and density that occur in the oil when no pigments are present. Thus pigments alter the course of the drying process in un bodied linseed oil to an extent that varies for different pigments. It now appears that pigments similarly alter the drying of bodied linseed oil, alkydresin vehicle, and phenolic-resin var nish. The increase in density during drying was always less than that cal culated from the changes in the unpigmented vehicle by an amount that ranged from 0.01 to 0.21. The short age was usually greatest for white lead, antimony Oxide, and zinc oxide, and least for titanium dioxide and mag nesium silicate. Also, the shortage was usually greater in bodied linseed oil and phenolic-resin vehicles than in raw linseed oil and alkyd-resin vehicles. When soaked in water for 3 days, all films, dear or pigmented, decreased in density to an extent roughly propor tional to the quantity of water absorbed unless the absorption was less than 1 per cent. Exact proptionality is not to he expected because there was a loss of soluble substances from the films during soaking. On redrying after soaking, the lost density was regained or exceeded by all films except two--antimony-zinc and titanium-zinc in alkyd-resin vehi cle. Higher density after redrying could come from further aging or the paint during the soaking and drying periods, or from loss of soluble ingredients of low density. The two paints that failed to regain their lost density were also the two with the lowest swelling effi ciency, which may indicate that they acquired some internal voids from the soaking and redrying treatment. Absorption of Water: When the clear vehicles were soaked for 3 days, the amount of water they absorbed de creased in the following order: un bodied linseed oil, 21.8 per cent; bodied linseed oil, 9-0 per cent; alkyd resin, 7.7 per cent; phenolic resin, 1.5 per cent, since the film of alkyd resin was thicker than the others, the absorp tion by alkyd-resin vehicle may be as dreat or even somewhat greater than by bodied linseed oil vehicle. If all GLD38015 sorption to be expected if the vehicle absorbed proportionately as much water as it did without pigment. Thus in unbodied oil, where 15.3 per cent ab sorption might be expected, the anti mony oxide paint absorbed only 5.3 per cent. Basic carbonate white lead likewise reduced the absorption in un bodied and bodied linseed oil, but failed to reduce or perhaps slightly increased the absorption in alkyd-resin and phenolic-resin vehicles, although the absorption still remained relatively low. All other pigments greatly increased the absorption of water above that ex pected for the proportion of clear vehicle in the paint. Among the singlepigment paints, zinc oxide and mag nesium silicate caused the highest ab sorptions. Zinc oxide was highest in unbodied and bodied linseed oil and in alkyd resin, but magnesium silicate exceeded zinc oxide in phenolic-resin vehicle. In zinc oxide, the absorption reached 83.3 per cent in unbodiea lin seed oil and 61.8 per cent in alkydresin vehicle. Magnesium silicate in phenolic-resin vehicle absorbed 15-3 per cent. Absorptions by titanium di oxide paints were 34.9, 6.8, 16.6, and 2.5 per cent in unbodied oil, bodied oil, alkyd resin, and phenolic resin, respectively. Absorptions by mixed-pigment paints, all of which contained at least one high-swelling pigment, were rela tively high. Presence of a low-swelling pigment in the mixture moderated the absorption appreciably in bodied lin seed oil or phenolic-resin vehicle, as was previously found in unbodied lin seed oil (4), but in alkyd-resin vehicle, mixtures of low- and high-swelling pigments usually showed higher ab sorption than the zinc oxide alkyd-resin paint. Apparently, high-swelling pig ments tend to dominate low-swelling pigments in all vehicles. Swelling: The volumetric swelling of films was usually equal to or a trifle greater than the volume of ab sorbed water. The swelling efficiency was therefore usually 100 per cent or a little over, which means that each cubic centimeter of water absorbed increased the volume of the film by a like amount or a little more. The slight excess of swelling over absorption may be due to the experimental difficulty of weighing the soaked films in air without significant loss of absorbed water by evaporation. If some films really did swell by more than the vol ume of water absorbed, it may be that the films, after their initial formation and hardening (1, 14), acquired in ternal compression stresses from fur ther absorption of oxygen, and that the stresses were released to add an increment of volume as soon as ab sorbed water plasticized and softened 4 the films. The data show no evidence of the compression of swelling liquid reported by Weinmann (22). A swell ing efficiency greater than 104 per cent appears in Table 1 only when the absorption was less than 8 per cent and tne excess of swelling over absorp tion less than 1.3 percent; that Is, when the measured values were small enough to make the experimental errors relatively large. Nevertheless, the relatively large swelling efficiency of the clear film of alkyd-resin vehicle, for example, appeared again when the corresponding weathered film was tested. Similarly, a swelling efficiency of less than 97 per cent was recorded only for antimony oxide in bodied oil, for which the absorption was only 0.8 and the volumetric swelling 0.7 per cent. The two measurements of 97 per cent swelling efficiency, however, may be accepted as true deficiency in swell ing because the film density was less after redrying than before soaking, which indicates that soaking developed void spaces capable of holding free water, the evaporation of which caused no shrinking of the swollen films. All films swelled in area when soaked in water, but areal swelling was not proportional, in general, to volumetric swelling. Usually, areal swelling was much less than volumetric swelling, but sometimes the two were nearly or even fully equal, especially when the absorption was relatively low. Of course, when areal swelling was large in relation to volumetric swell ing, swelling in thickness was corre spondingly small, and for seven of the films it was even negative. That is, the films became thinner on swelling. It was shown previously (3, 7) that th; ratio of length to width in areal swelling of linseed oil paints may vary greatly among paints of different pig mentation, and may vary greatly for any one paint according to the too! used and methods followed in applica tion, particularly if the paint contains acicular pigments that may become preferentially oriented when the paint is spread and dried. Similar variations were observed among paints in the vehicles of the present study, but these variations need not be presented in detail. It is sufficient to point out that volumetric swelling is independent of a number of considerations that mark edly affect the distribution of swelling among length, width, and thickness (11). Volumetric swelling is therefore more readily reproducible and, in general, more reliable than areal or linear swelling. Shrinkage on Redrying; On redry ing, all swollen films shrank to a volume less than that before soaking, except the film of antimony-zinc paint in alkyd resin, which shrank only to its volume'before soaking. Such loss in volume is due primarily to loss of soluble substances during soaking. Since the nature, and therefore the density, of the dissolved substances has not yet been determined, it is not possible to compare the volume lost by the film with that of the substances dissolved. However, another factor must also be considered, namely, the change in density of the film before soaking and after redtying due to aging of the film or to development of voids within die film. The antimony.zinc paint in alkyd resin, which had the same volume after redrying as before soaking, despite a solubility of 3 3 per cent, exhibited such formation of voids by decreased density and low swelling efficiency. Similarly, titanium-zinc paint in alkyd resin, which showed a loss in density and low swelling efficiency, experi enced only 2.5 per cent loss in voiume despite 7.3 per cent solubility. After redrying, 27 of the 34 films remained larger in area than they were before soaking, 2 were unchanged in area, and 5 decreased in area. Of course, failure to shrink in area when the volume shrank caused correspond ingly large shrinkage in thickness. All seven films that were no larger in area after redrying than before soaking were either dear films or films pig mented only with white lead or with antimony oxide. When high-swelling pigments were present, the redried area was always greater than the pre soaked area. Loss in Weight While Soaking: If allowance is made for the differences in film thickness, the solubilities of the films of clear unbodied and bodied lin seed oil and alkyd-resin vehicle may be considered about equal at 5.8 grams per 100 cubic centimeters for 3.7-miI thickness. The phenolic-resin varnish was much less soluble, 1.3 grams per 100 cubic centimeters. For the 70 per cent by volume of vehicle in the pig mented films, the solubility, if the pig ments had no effect, should be 4.1 grams per 100 cubic centimeters for the first 3 vehicles and 0.9 gram per 100 cubic centimeters for the phenolicresin varnish. In unbodied linseed oil, the solubil ities for paints with pigments other than zinc oxide were not far from 4.1 grams pet 100 cubic centimeters, but zinc oxide raised the solubility to 5.9 grams per 100 cubic centimeters. In bodied linseed oil, no paint was higher than 1.9 grams per 100 cubic centi meters in solubility. In alkyd resin, the solubility of all paints was more than 1.9 grams per 100 cubic centimeters, but only the zinc oxide and the titanium-zinc paints were much more soluble than 4.1 grams per 100 cubic GLD38016 absorption by the pigmented films took place in the vehicle, without any effects from the pigments, the absorption should have been 15.3 per cent for all films with unbodied linseed oil, 6.3 per cent for all films with bodied oil, 5.4 per cent for ail films with alkyd resin, and 1.0 per cent for all films with phenolic resin. Pigmentation, however, resulted in higher or lower absorptions, according to the nature of the pigments. To see whether absorption of water by pigments in the absence of vehicle could account for the effects of pig ments on absorption by paint films, 3- to 4-gram samples of white lead, zinc oxide, titanium dioxide, and mag nesium silicate were each soaked in 150 cubic centimeters of distilled water for 3 days. The water was then de canted and the pigments allowed to dry at room temperature until the weight was constant. The white lead gained 0.044 per cent by weight or 0.3 per cent by volume, zinc oxide gained 0.48 per cent by weight or 2.69 per cent by volume, titanium dioxide gained 0.023 er cent by weight or 0,09 per cent y volume, and magnesium silicate lost 0.18 per cent by weight. Thus the mag nesium silicate contained a small pro portion of water-soluble matter and the other pigments retained slight amounts of water that was presumably adsorbed. Zinc oxide, as was to be expected, retained more water than white load or titanium dioxide, but the amount was far too small to account for the large effect of zinc oxide on the absorption of water by paint films. With paint films of any one pig mentation, the absorption was nearly always highest for the unbodied lin seed oil vehicle and lowest for the phenolic-resin vehicle, and was always higher for the alykd-resin vehicle than for die bodied linseed oil vehicle. The differences attributable to kind of vehicle, however, proved less than dif ferences caused by kind of pigment. Antimony oxide, in all four vehicles, reduced the absorption below the ab- Table I.--DATA FOR FREE, UNWEATKERED FILMS OF DIFFERENT Oil OR VARNISH VEHICLES WITH AND WITHOUT FIGMENTS--CHANGES IN DENSITY! ABSORPTION OF WATER AND SWEUING IN VOLUME, AREA, AND THICKNESS WH5N SOAKED IN WATER FOR 3 DAYS. SWELLING EFFICIENCY, SHRINKING IN VOLUME, AREA, AND THICKNESS WHEN REDRIB), AND LOSS IN WEIGHT BY LEACHING DURING THE SOAKING FERIOD Vehicle Density of DCQVOlltlle in liquid point Density of fill Dry Swollen Bedrlcd in wster 1t Initial xAbsorptlcax VwsUthi 1.1 i IwUlai : MrM state it of film : 7. 1 A/fn 1 Zb tin *r*t Zb i :v s 1u m: A 9, lUileEatMi x AY, t 1 AT, t , j Vo Imbs x Are* xlhlekaesstushinc 1 AY, 1 at, 1 47, ! B2/V0 it! : Mil* s Percent s ftr- tPercent: Percent i Percent s Per* iMrewti Percent tSr. per x i cent t t t i em : : wSOcci tlneeed oil, not bodied Linseed oil, beet bodied Alkyd vehicle Phenolic-resin vsrnish 0.93 .97 1.00 1.02 tlneeed oil, not bodied Linseed oil, hest bodied Alkyd vehicle Phenolic-resin vsrmsb 2.70 2.71 2*79 2.76 CLEAR rXRICLRS CdRUXIM FO MOOFT l.U % 1.11 : 1.11 : l.U : 1.09 1.09 1.10 1.10 x 1.12 : l.U : 1.12 : l.U * : t i 3.1 3.6 5*6 3.9 21.8 9*0 7.7 15 * 22.0 > 9.8 t 9.2 s 6.1 1 8.9 1 6.0 l 1.7 2 1.0 *1 10.8 t 1.4 % -.7 t -3 t 101 102 116 013 BASIC CAKMKMt WHW HAD, 8DKU-PKBCKT VAXKB, riant VOUM 0.30 2.frj t 2.62 ; 2.91 : 2.80 : 2.74 2.78 2.00 2.79 : 3.90 : 3.87 : 3.98 : 3.83 ! t ! 3-8 4.8 4H.8 8.9 t 9.3 i 2.1 3*1 2 4.2 : .6 9-7 : 10.0 : 2.7 1.6 ; 2.3 .3 5.3 t 1.0 1 3.8 t .6 1 104 135 % t 6.o 1 -4.9 : x *4.9 1 -3.6 ! : -4.0 t *3.1 t : -1.3 1 -.1 ! -1.3 -2.2 -5.7 -1.3 : -2.5 1 t -2.3 1 t -3.3 1 t -1.2 : *.l : 0t .3 0i -2.1 -2.5 -3.3 -1.2 5.8 3.1 3.3 1.3 3.8 1.9 3.0 .6 Linseed oil, not bodied Linseed oil, best bodied Alkyd vehicle Phenolic-resin vsrnish 2.38 8.39 2.46 2.44 Linseed oil, sot bodied Linseed oil, best bodied Alkyd vehicle Phenolic-resin vumish 1.91 1.93 2.05 1*97 Linseed oil, not bodied Linseed oil, hest bodied Alkyd vehicle Phenolic-resin vsrnish 2.33 2.35 2.41 2.39 Linseed oil, not bodied : Linseed oil, best bodied: Alkyd vehicle : Phenolic-resin vsrnish : 1.51 1.53 1.57 1.57 utroKtct acme , smau-nawn mm, naan van* 0.30 2.54 : 2.42 t 2.57 i 2.47 : 2.52 2.42 2.56 2.48 : 2.61 t 2.42 : 2.61 : 2.50 : r : : 3-4 5.8 3.3 5.3 >.3 x 3*8 * 3-9 .8 * .7 * .7 3.3 i 2.7 1 1.8 .9 : 1.2 1 1.3 3 t -.5 : -.9 j -3.0 1 KCTCLX TltWWM DltKIDt, SIBOU-PICMBT PADRS, PlOHDPt VOUJMB 0.30 2.06 ; 1.99 s 2.11 ; 1.99 s 1.80 1 2.U : 1.94 : 2.02 r 1.97 : 2.15 . 1.97 1 2.00 1 3-5 5.3 3.4 6.6 34.9 6.8 16.6 3.5 : 35.5 * 1T.1 1 1 7.7 t 5*7 x 17.2 1 10.3 : 3.2 \ 2.5 12.4 0 3.5 .5 znc an*, w mo u -k o o b r mm nmmn vetna 0.30 t * 3 2,44 3 2.45 : 2.55 : 2.44 : 1.80 2.10 1.96 2.32 i 3.47 : 3.46 : 3.58 -. 3.45 1 : i - 3-7 4.3 3-0 5.0 83.J 31.8 61.8 9.8 t 81.6 t 38.4 t 31.5 I 10.9 t 61.5 t 31.5 l 9.9 > 4.7 t 39.3 1 17.3 : 89.8 1 4.0 X Munsiw s il ic a t e , siML*-naan rum, j immu v c u m o .j o : 1.68 t 1.44 : 1.70 t 2-9 s 56.* : 56.1 13.7 37.3 1 : 1.60 : 1.47 i 1.62 : 5.2 x 39.0 i 86.5 , 14.0 t 10.7 1 : 1.68 : 1.49 x 1.69 s 3.2 : AO.) : 40.0 : 16.1 < 17.6 a : 1.62 : 1.54 : 3.63 : yi t 15.3 1 .7 1 7.1 1 6.3 t 109 87 133 133 103 113 103 138 96 99 100 101 100 102 100 102 : -4.3 : : -.7 t 1 -2.4 x t -1.6 x -1.4 1 *.2 I -.7 ! *S ! -2.9 *.9 -1.5 -2.0 4.7 .6 2,4 1.3 t -4.2 t *3.3 t -2.2 i .3 J : -2.9 t .4 . 1 -1.2 : .4 1 -7.1 -2.1 -3.2 -1.3 4.1 1.3 3-0 .3 t -3*6 t 1 -1.0 t t -3.6 t 1 -1.3 : 5.0 t *3.0 t 6,0 : 1.3 1 -8.0 -4.0 .9.2 -2.6 5.9 1.6 6.4 1.6 t -3.6 : *3.3 , 1 -2.1 \ *1.7 t -3.6 1 3,8 ? -1.4 t .7 -5.8 1 4.5 -3-8 ! 1.7 -S.3 : 4.5 -1.9 1 .9 Lineeed oil, be*t bodied: Alkyd vehicle : Ph<nolic-r*ein vsrnish : 2.63 2.69 2.67 Linseed oil, best bodied: Alkyd vehicle : PbeaoLfe-resin. vsrnish x 2.30 2.44 2.U iz a d -z h c , ians>-mcan nuns 123$, r o o t y c u m 0.30 .2.72 ! 2.48 j 3.74 4.5 : 16.3 i 16.5 1 5.9 9.0 2.82 2.06 : 3.84 t 3.1 ! 73-5 1 72.6 1 36.9 | 94.4 3.73 ! 2.6T : 3.74 t 4.9 1 4.5 t 4.7 t 3.8 t .8 jarnMarr-M*c, tanm-naanr mm az& , naan num 0.30 .3.4$ 1 2.34 : 2.47 3.57 : 1.98 : 2.5/ 5.3 3.1 1 t 18.0 64.3 1 18.1 1 8.8 t 8.1 1 63.3 ( 34.0 t 21.2 3.46 : 3.38 : 2.48 t 5.0 s 7.3 : 7.5 ! 4.6 l 1.4 ntAUuK-znc, nxo-naan mu t s n^, naan v o u m 0.30 x t t t x : 101 99 104 103 97 103 x -1.0 1 .9 * -1.8 1 .9 1 -1.9 t 411.5 X -12.2 1: 3.8 1 -.6 s +.4 i -1.0 11 .5 t -1.0 x 1.5 * -2.5 11 .5 t 0 t *13.4 x -11.5 1 3.3 x -1.3 t .7 x -2.2 1: 1.3 Linseed oil, best bodied: Alkyd vehicles : Phenolic-resin vsrnish : 3.04 1.85 2.09 2.11 x 1.90 ; 2.12 : 4.8 ,1 33.3 t 23.3 J 11.0 t 10.1 1 101 1.97 * 1.52 s 1.95 t 3.3 !: 86.4 1 84.0 1 36.4 , 33.0 t 97 2.11 : 2.05 : 2.13 * 5.3 : 6.4 i 6.5 > 3.9 1 -.8 t 102 1 .1.1I *3.4 , -3.3 ! .9 , -8.51 *15.8 , -13.3 1 7.3 .1.6: *1.6 1 -3.3 1.8 YTEAXIUK-AXnMORT-HLORESrJM SILICATE MUTT **25*57, F"** VttlW 0.30 Alkyd vehicle : 1.86 8.03 : 1.84 ! 3.04 , 3.0 i 83.8 1 33.4 1 8.1 1 11.5 103 -J-* t *.8 1 -3.6 1 3.4 Xlhc r* >i..... L. tbu p*lnt ccotained dioxide, sine oxide, xod ad^Mdiua elllcete; paint t /d v 0.30. 4 GLD38017 centimeters. In phenolic resin, no paint reached 1.9 grams per 100 cubic centi meters in solubility, and only the zinc oxide, titanium-zinc, antimony-zinc, and antimony oxide paints exceeded the solubility expected of the dear vehicle, 0.9 gram per 100 cubic centi meters. There seems to be a rough relation between the effect of pigments on the solubility of paint films and on film absorptions and swelling (7). Meier and Schulz (16) found that removal of water-soluble ingredients from pig ments reduces the swelling of paints made with them, but that paints with some pigments, particularly zinc oxide, still swell more than the clear vehicle, even if the pigments have been thor oughly leached. It may well be that pigments affect film swelling at least in part by altering the formation of water-soluble substances formed from the vehicles during oxidation, aging, or hydrolysis (1, 15). Weathered Films Table 2 gives the effects of weather ing on oil-type paints that were weath ered artificially for 15 days (1,200 kilowatt-hours of electric input), and the subsequent effects of soaking on these artificially weathered films. Loss in Weight While Weather ing: All films lost weight during the 15-day period of artificial weathering, as other workers have found (1, 13). Among the dear vehicles, the losses per 100 cubic centimeters were 28.1 grams for unbodied linseed oil, 12.7 grams for bodied linseed oil, 6.1 grams tor alkyd-resin vehicle, and 2.3 grams for phenolic-resin varnish. For the proportion of vehicle in the pigmented films, the losses would be 19.7, 8.9, 4.3, and 1.6 grams per 100 cubic centi meters, respectively, if the pigments exerted no effect. In unbodied linseed oil vehicle, all pigments except zinc oxide made films that lost more than 19-7 grams in weight while weathering. Zinc oxide decreased the loss materially. In bodied oil vehicle, on the other hand, all pig ments except magnesium silicate re duced the loss to less than 8.9 grams. Zinc oxide or any pigment mixture that included zinc oxide was particu larly effective in reducing this weight loss. Magnesium silicate increased the loss slightly. In alkyd-resin vehicle, pigments usually increased the weight loss to more than 4.3 grams, but zinc oxide and mixtures containing zinc oxide gave only slight increases or even slight decreases in the weight loss. In phenolic-resin varnish, pigments usu ally increased the weight loss, and zinc oxide increased it as much or more than other pigments. Since water extracts appreciable quantities of zinc oxide from films of linseed oil paints, but extracts little or none of the other pigments used in these tests (3, 7, 12), it might be expected that the loss of weight on weathering would be higher for the paints that contain zinc. The fact that zinc oxide tends instead to reduce the weathering loss indicates that the effect of zinc on the solubilization of the vehicle is more important than the re action of zinc with arid decomposition products of the vehicle to form soluble soaps (1, 15). Density: All films were higher in density after the weathering period than they had been when dried For 10 days but not yet weathered. The in crease was due to further contraction in volume on aging and weathering and to loss of vehicle constituents that are much less dense than the pigments. When dear films of the four vehi cles were weathered, the density in creased 5.4, 4.5, 3.6, and 4.5 per cent for unbodied oil, bodied oil, alkyd vehicle, and phenolic-resin varnish, re spectively. For the films of unbodied oil paints, the density increase was 7.8 to 15.7 per cent, and was always greater than that for the clear oil without pigment. For the films of pig mented phenolic-resin varnish, on the other hand, the density increase was only 2.5 to 3.5 per cent, and was always less than that for the clear var nish. For the films of bodied oil paints and alkyd-resin paints, except those containing zinc oxide, the density in crease always exceeded that of the dear vehicle. The increase was 6.0 to 7.5 per cent for bodied oil paints, and 4.2 to 7.8 per cent for alkyd-resin paints. When the paint contained zinc oxide, the density increase was always less than that for films containing any other pigment in the same vehide. In general, pigments exerted an effect on density increase during weath ering that was opposite to their effect , during film formation and drying. Pig mentation delayed the density increase during drying of the liquid paint, but when the pigmented films were weath ered, they tended to catch up with the increase in density of die dear vehicles. A comparison of die density of the weathered films with the density of 10-day-old films calculated on the assumption that pigments do not alter the behavior of the vehide shows that weathered films of unbodied oil paints reached a density 2 to 12 per cent reater than theoretical for 10-day-old lms, weathered bodied oil or alkydresin paints came within 2 per cent above or below theoretical, and phe nolic-varnish paints failed to reach theoretical by 1 to 4 per cent. When the weathered films were soaked in water for 3 days, the density decreased roughly in proportion to the quantity of water absorbed, unless the absorption was less than 2.5 per cent. After redrying, the density returned to its value before soaking, or slightly higher, for all but seven films, sue of which became somewhat porous, as indiuted by their lower swelling efficiency and larger volume after redrying. Absorption of Water: Weathering had little effect on the absorption of water by dear films of unbodied lin seed oil and of phenolic-resin varnish. It diminished the absorption by clear films of alkyd-resin vehicle only slightly, and greatly increased the ab sorption by dear films of bodied lin seed oil so much that the bodied oil films absorbed more water after weath ering than the unbodied oil films did before they were weathered. Pigmented films that absorbed more than 9-0 per cent of water before weathering always absorbed less water after they were weathered. They ab sorbed much less if their absorption before weathering exceeded 20 per cent. Weathering also diminished the absorption of most pigmented films that absorbed lei duui 9.0 per cent before weathering. The exceptions were white lead in unbodied and bodied oils, antimony oxide in bodied oil, alkyd resin, and phenolic resin, and lead-zinc in phenolic resin. For these exceptions, weathering increased the absorption slightly or wiled to decrease it. In weathered films of unbodied lin seed oil, the only pigment that caused absorption as great or greater than that to be expected of the proportion of dear vehide in die film was zinc oxide. In weathered films of bodied oil, dear films were so much more absorptive after weathering than before that all pigments reduced the absorption below that expected of the proportion of vehide in the films. In weathered films of alkyd resin or phenolic resin, how ever, tiie absorption was as great or greater than that expected of the pro portion of vehide In the films, except with antimony oxide, with which the absorption was slightly less. Swelling: For nearly half of the weathered films, the volumetric swell ing was less than the absorption of water, and the swelling efficiency was therefore less than 100 per cent. That was the case with all unbodied oil painty although the unbodied oil with out pigments had 100 per cent swell ing efficiency. The tendency for weath ering to reduce swelling efficiency has been observed repeatedly in previous work, and is attributed to the develop ment of voids within the film that are capable of holding free water that causes no swelling. GLD38018 5 Weathered films of bodied oil were the only unpigmented films to have less than 100 per cent swelling effi ciency. Bodied oil paints that contained white lead had only 75 per cent swell ing efficiency, and the zinc oxide and lead-zinc paints had a 99 per cent effi ciency. Bodied oil paints with antimony oxide, titanium dioxide, or magnesium silicate had at least a 100 per cent swelling efficiency. Weathered films of unpigmented alkyd resin, like the cor responding unweathered films, swelled beyond the volume of water absorbed, which may indicate a release of inter nal stresses acquired when the films originally dried. If such a release did occur, these internal stresses were ab sent or less significant in the films of weathered alkyd-resin paints, none of which had much more than 100 per cent swelling efficiency, and three of which had less than 100 per cent effi ciency. Weathered films of phenolicresin varnish, with or without pig ments, were too low in absorption and swelling to permit reliable deductions from the variations in swelling efficiency. The swelling in area and thickness of weathered films exhibited much the same variations from proportionality to volumetric swelling that nave already been discussed for unweathered films. Shrinkage on Redrying: On redtying after soaking, all but one of the weathered films shrank to a volume closer to that of the film before soak ing than was the case with the un weathered films. For the one exception, the redried volume was within 0.7 per cent of the initial volume for both unweathered and weathered films. As a rule, the redried volume was less than the volume before soaking, but for weathered films of six paints, the redried volume slightly exceeded the initial volume. The smaller residual shrinkage of weathered films may be attributed to less solubility during soaking, and perhaps to less swelling during the soakingperiod. On redrying, weathered films, like the unweathered films, tended to 1 fobk 2.--DATA FOR FREE, WEATHERED FILMS OF DIFFERENT OIL OR VARNISH VEHICLES WITH AND WITHOUT PIGMENTS--CHANGES IN DENSITYi ABSORPTION OF WATER AND SWELLING IN VOLUME, AREA, AND THICKNESS WHEN SOAKED IN WATER FOR 3 DAYS. SWELLING EFFICIENCY! SHRINKING IN VOLUME, AREA, AND THICKNESS WHEN REDRIEDi AND LOSS IN WEIGHT BY LEACHING DURING THE SOAKING PERIOD Weight lose Density of film lt t Initial (Absorption: Swelling tt i Swelling Retried :Lo m in Vehicle weathering Dr, : dvollea: Redried: of film BlAo in vmtert : *a kfs9 1 la iXa arta; In *Vo :valuw: AR. :thlcknt: : t . t AT* 1 A sVftluM: Area :lfclcknee*:*oaking t AY, : AOr 4r s * Percent ; MU* Percent : Per* :Percent: Percent 1 Percent s cent s'"" " s 1 t Per* j Percent: cent 2 Percent `S10i 0 CC. Llneeed oil, not bodied? Llrvcecd oil, h*at bodied Alkyd vehicle Menoltc-reeln vamieh 3.1 12.7 6.1 > } Linseed oil, not bodied llneeed oil, heat bodied . Alkyd vehicle Pheooilc-reeltv vamlah 271..60 4,9 2.1 Linseed oil, not bodied Linseed oil, heat bodied Alkyd vetaleie t Pheaolle-reeln varnish 36.6 115..03 2.9 Llneeed oil, not bodied Llneeed oil, heat bodied Alkyd vwhlei* Pheaolic-reela varnish 27714....4431 Llneeed oil, not bodied Llneeed oil, heat bodledi Alkyd vehicle Ffcenollc-reeta varnish 11.T 6J3...1J6 Llneeed oil, not bodied Llneeed oil, heat bodied 4Alkyd vehicle J"b aollc*reeta vamleh 88.6 1101..56 2.1 Ltnaeed oil. heat bodiedt Alkyd vehicle ftienollc-reela varnish 162...663 CLEAR retXCJJB COMEADUBO HO FXGMOT 1.17 ! 1.16 : 1.15 : 1.16 : l.l4 1.13 1.15 1.16 : 1.17 : 1.19 : 1.15 : 1.1k ! t ! : 3-7 4.1 5.1 3.9 20.4 26.1 6.8 1.5 t 80.5 t 6.6 2 27.S : 10.1 5.5 < 3-5 1 2.0 : 1.0 t : t : 15-5 * 16.9 ! 1.0 t 02 100 98 114 133 SASIC CARBOKAff WIR LKAJ>, KPXE-PIOMBf? MOTS, FIOKBfT VOUW 0.30 1.26 ! 2.99 : 1.05 : 2.&T : 3.06 2.92 a.96 2.85 : 3.U 1 2.96 > 3.06 : 2.68 s s 1 ; 2.3 4.0 42..79 9.9 8.7 .7 1 6.8 : J.6 s 1.0 t 4.2 : 6.3 1 1.6 : 1.5 : 1.3 2 0 s 8.8 i 3.8 ! 2.7 .9 88 75 102 100 AxriMtirr o o d k . uMU-picMrr aurra, naan v o u m 0.30 2222..,.5769670k : : : : 2222....59576b78 ! : J ! 2222....56795177 t : r J 42..39 63.-61 122...455 .9 s 1 2 122...532 t t 122.,.313 1 1 2 1 .71 .9 > -011-.,.700 1 t t > 19020 796* mu trantM sic*in*, snai-naan Mam, namt nuu 0.30 2.36 : 22..2162 : 1 2.06 : 22..3102 22..1086 : 2.17 11 1 222...1206?3 : i : t 34-.21 ?* 6.5 .83..39 31.A9 t ( > 1 5233..-.7019 t t t 2 1312....6969 ! 1t t 11..09 *1 1-..43 2 0 2 0 3ZBC K DC, SIBOU-PIOMOT MOTS, PMORT VOUMI . 2222....5656532] : s > J 222,..33bO06 2.65 t t : : 2.61 22..5652 2.S3 t ! 1 : 632..-838 6.3 1246.,22 17.6 2 2 1 112646..,600 t t s 743*..446 1 t t 86.0 U10..93 1 5.3 > 5.5 1.3 1 3.7 n1960o109 92 99 & taamxm tzucAts stMu-PioMr mu m, naan vouw 0.30 1111....77661569 1 : t 11..861l 1 6I.69 .l 1 ! 1 t 1111....68777892 1 I 1 s 6235...2641 11962...199 8.9 t 1 ! t 119929-...132811s 665...630 2.0 t 1 1 6.7 1 7.3 65.6 1 .S 1 99 110012 105 LiAD-tnr KnsD-pxoMBR Mknrm u,;( lien veuMB 0.30 2.82 1 28..9726 t -. 8.58 22..6690 1 2.82 8 7: 2.93 : . t 1 * 6.5 16..06 1205..37 5.6 1 t 1850*.38 1 > 54-.16 t > t 5.6 t 8.1 i 9.9 > 123..46 1 1 1990980 2.2 t -.7 1.4 * 2- 1.0 2 -1.7 1 2.1 2 -.3 ,S 1 *0.1 1 .2 s 1.8 1.2 t .5 s1.0 .5 41 2.3 : 2 . i .6 1 2 -T 1 -.7 > -.6 1 ,6 t -.5 2.1 1 2.9 1.0 2 2.0 0 ; 1.6 -.2 2 l.l 222 --1....64571ti -1--....9620 s s : *aO2 2 s 11..05 .6*.7 2 .3 2 .9 -1--...458 1 ! x -.7 --0111...313 t > s O1.0 1-..93 221 * 1...840 .3 0 .1 > x .3 -.71 1.4 2 --...61311 0-0.3 -.5 1 t 2 *...6385 -122...102 l t 1 -.9 > -13..06 1 2 +..152 -2.a1 -21..52 2 2 3..66 .93 1.3-.31 . * -.9 * > 1.21 .* 1 1.1 --23..33 1.29 1.6 a it o o r -u k / k bc b >-p io n *t mm uof, mo o r * *u u mr 0.30 Llneeed oil, heat bodiedt Alkyd vehicle Phenollc*reeia varslch 1.6 11..97 288...556656 1 : j 222...163775 1 1 222...655553 1t 1 652...609 21406...272 I 14.0 t ! t1 20.3 * 6.5 95*.36 1.6 t i t 982...146 >t1 95 100 197 -.7.t t t -.71 -3..49.1s> - 13-...708 1.0 .5 .7 nram-tiK, k k a-pj o n o n Hum namn t c u mi 6.36 Llneeed oil, heat bodied tAlkyd veblelai Meaollc-recia vmmieb 1.2 M2.3 t 1* 22..0129 2.16 1 s 1 122...810522 !2 22..1072 s 2.17 > t 1 6.6 6J..7* 1206..21 6.3 t t1 81695...819 > >s 866...095 1 1 t lt8.O.T 1 1 1.9 1 11910190 I Ig -..66.7%81 13-...382 1 a -3"..8I8 ...735 I nvumM-Junmcm-MuauuK r j u c as h u r t *j j Zj t . namn t c u m c.s vMtcl, 7.6 2.07 1 2.06 i 2.10 t 3.1 6.6 t 4.6 2 2.2 t 2.0 t 100 2 -1.5 f .9 * 7 .6 w plemt la tUa patat coatalaa* titaatw tloaMa, Mae eM4a, aa ipeili etlleatej (Mat tBtJBjl / O.JO. c GLO 38019 shrink less in area and more in thick ness, rather than in proportion to the volumetric shrinkage. The discrepancy generally was less for the weathered than for the unweathered films, how ever. Thus only 10 of 34 weathered films remained larger in area after redrying than before soaking, com pared to 27 of the 37 unweathered paint films. Weathered paint seems to be less able to readjust the ratios among length, width, and thickness during changes in volume than is unweathered paint. Loss in Weight While Soaking: The solubility of most weathered films during soaking was less, usually much less, than that of the corresponding films before weathering. Exceptions were four bodied linseed oil paints and two phenolic-resin paints, among which the solubility of the weathered film exceeded that of the unweathered film by more than 0.2 grams per 100 cubic centimeters only when the solubility of the weathered film was no greater than 1.4 grams per 100 cubic centi meters. Among the weathered films, the solubility was always higher for un bodied linseed oil films than for the corresponding films with any other vehicle. Bodied linseed oil films stood second in solubility unless they con tained zinc oxide, in which case the corresponding alkyd-resin films usually were more soluble than the bodied oil films. The solubility of weathered phenolic-resin films was less than 1.0 gram per 100 cubic centimeters except when the pigment was white lead or lead-zinc, for which the solubilities were 1.1 and 1.4 grams per 100 cubic centimeters, respectively. Conclusions The formation and hardening of films results in increased density, and the density increases still further when the films are weathered. Ten-day films of all four of the clear vehicles had densities of 1.11. After weathering artificially for 15 days, the densities ranged between 1.15 and 1.17, although the densities of the liquids before spreading varied from 0.93 to 1.02. When pigments are incorporated in the vehicles, the density increase dur ing formation and hardening of films is diminished. The increase in density during drying is always less than that calculated on the assumption that the vehicle in paint undergoes the same changes in weight, volume, and den sity as the clear vehicle without pigment. Pigments alter the course of drying of bodied oil and of phenolic-resin s'ehides relatively more seriously than they do that of unbodied oil and alkydresin vehicles. White lead, antimony oxide, and zinc oxide exert more effect than titanium dioxide and magnesium silicate. During weathering, however, pigments have an opposite effect be cause the density increase of pigmented films tends to catch up with that of the corresponding clear vehicle. The four dissimilar vehicles resem ble one another in that their films absorb water, swell, and lose soluble matter when soaked in distilled water for 3 days and, on redrying, shrink to a volume usually smaller than that before soaking. The changes are usu ally much the greatest in magnitude for unbodied linseed oil (21.8 per cent absorption), least for phenolic-resin varnish (1.5 per cent absorption), and about the same intermediate magnitude for bodied linseed oil and alkyd-resin vehicle. Weathering leaves the changes nearly as great for unbodied oil and for phenolic-resin varnish, reduces them -slightly for alkyd-resin, but in creases them for bodied oil to as much or more than those for unbodied oil. Pigmentation greatly alters the ab sorption of water by films of all four vehicles. Even the low-absorbing phe nolic-resin varnish, which absorbs only 1.5 per cent when the film contains no pigment, absorbs 15.3 per cent when pigmented with magnesium silicate. Absorptions greater than 83 per cent in unbodied oil films, 31 per cent in bodied oil films, and 86 per cent in alkyd-resin films can occur with some pigments. Although weathering mate rially decreases the absorption by pig mented films, it may remain well above that of the vehicle without pigment. Among the pigments, zinc oxide, magnesium silicate, and titanium di oxide increase absorption in all vehi cles, but antimony oxide and white lead reduce absorption or at least fail to increase it very much. In mixture, high-absorptive pigments dominate over low-aosorptive pigments. Unweathered films, either without pigments or pigmented at 30 per cent pigment volume, usually swelled in volume by the volume of water ab sorbed; that is, their swelling efficiency is 100 per cent within the experimental error. There are perhaps rare cases of more than 100 per cent swelling effi ciency that maybe attributed to inter nal compression stresses from absorp tion of oxygen after initial formation and hardening of a film. Release'of these stresses adds slightly to volume as soon as absorbed water plasticizes and softens the film. Swelling efficiency less than 100 per cent, which is sometimes exhibited by unweathered films containing zinc oxide and more generally in weathered films, is attributed to development of porosity when soluble material is leached from the film. When redried. however, the film fails to shrink enough to close the voids. Volumetric swelling is distributed amsotropically and not very reproducibly between area and thickness of both pigmented and unpigmented films, both before and after weathering. When the films are redried after soaking, nearly all unweathered films and most weathered films shrink to less than, the volume before soaking because of loss of soluble material during soaking. Similarly, the density of the wet film decreases in proportion to the water absorbed. When the film is redried the density usually exceeds the density before soaking, because the soluble materials lost are relatively light in weight. Failure of a redried film to shrink to its volume or less than its volume before soaking, which is usually coupled with failure of the redried density to exceed the density before soaking, is taken as an indica tion of development of porosity within the film. All films lose weight while weather ing. The unpigmented films of the 4 vehicles lose weight in the decreasing order, unbodied-oil, bodied oil, alkydresin, phenolic-resin. Pigments usually tend to increase the weight loss by unbodied oil, alkyd-resin, and phenolicresin films, but they tend to decrease the loss by bodied oil paints films. Zinc oxide, however, reduces the weight loss or fails to increase it as much as other pigments in all vehicles except phe nolic-resin, in which zinc oxide causes greater loss than other pigments. Literature Cited 1. Bell, S. H. 1955. The structure of paint films. Jour, of Oil and Color Chemists Association Vol. 58, pp. 395-623. 2. Brasher, D. M. and Kingsbury, A. H. 1954. Electrical measurements in the study of immersed paint coatings on metal. 1. Comparison between capaci tance and gravimetric methods of esti mating water uptake. Jour, of App. Chem. Vol. 4, part 2, pp. 62-72. 3. Browne, F. L. 1953. The absorption of water, swelling, and solubility of free films of paint. Jour, of FPRS. Ill (5): 108-24. 4. Browne, F. L. 1954. Swelling of paint films in water. II. Absorption and vol umetric swelling of bound and free films before and after weathering. Jour, of FPRS, IV (6): 391-400. 5. Browne, F. 1. 1955. Swelling of paint films in water. III. Absorption and vol umetric swelling of bound and free films from air of different relative humidities. For. Prod. Jour. V (1): 92-6. 6. Browne, F. L. 195>. Swelling of paint' films in water. IV. Effect of thickness of film and pigment volume of paint. For. Prod. Jour. V (2): 142-6. 7. Browne, F. L. 1955. Swelling of paint films in water. V. Effects of different pigments. For. Prod. Jour. V (3): 192-200. 8. Bundies, J. 1953. Volume relations in paint coatings. Deutsche Farben Zeitung Vol. 7, pp. 177-8. 7 GLD38020 9. Buscr, K. 1954. Is the moisture-swelling of rust-preventing paints a measure for evaluating their quality? Deutsche Farben Zeitung Vol. 8, pp. 471-2. 10. Calkins, G. D., Pobereskin, M., Young, V. E. and Nowaclci, L. J. 1955. Tritium determines moisture gradient in attached protective-coatings. Nucleonics Vol. 13, pp. 76-7. 11. D'Ans, J. 1935. Research on coatings, particularly on film swelling. Farbe and Lack Vol. 61, pp. 54-912. Dunbar, R. E. 1951. Chemical changes in films with aging. Official Digest, Federation of Paint and Varnish Pro duction Clubs. No. 323, pp. 857-60. 13. Kaiser, Ewin B. and Couliette, J. H. 1951. Hardness, abrasion resistance, and accelerated weathering tests on pure pig mented and unpigmented paint-vehi cles. Official Digest, Federation of Paint and Varnish Production Clubs. No. 322, pp. 724-39. 14. Mandelkern. Leo and Long, F. A. 1951. Rate of sorption of organic vapors by films of cellulose acetate. Jour, Polymer Science Vol. 6, pp. 457-6915. Mayne, J. E. O. 1952. The protective action of paints. Research Science and Its Applications in Industry. V (6): 281 16. Meier, K. and Schulz, G. 1953. Pig ment electrolyte content in relation to film swelling. Farbe und Lack Vo!. 61, p. 220. 17. Mikusch, J. D. and Mebes, K. 1953. On the swelling behavior of conjugated and isomerizea oils. Farbe und Lack Vpl. 61, pp. 9-16. 18. Park, W. R. R. 1955. The water resist ance of air-dried alkyd resins. Official Digest, Federation of Paint and Varnish Product Clubs. No. 365, pp. 371-86. 19. Singer, Rudolf J. R. 1954. she swelling of oil films in water in relation to the composition of the oils. Farbe und Lack Vol. 60, pp. 189-94, 261-3, 298-304. 342-30, 391-7. 20. Van Loon, J. 1953. (Cited by S. H. Bell, reference 1). Vertkrooiek Vol. 26, P- 18. 21. weber, F. 1952. Swelling of red lead and lead cyanamide paints. Deutsche Faiben Zeitung Vol. 6, pp. 288-92. 22. Weinmann, K. 1934. Swelling of im mersed coatings. Farbe und Lack Vol. 60, pp. 545-51. GLD38021